Tailoring the Energy Landscape in Quasi-2D Halide Perovskites

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Communication

Tailoring the energy landscape in quasi-2D halide perovskites enables efficient green light emission Li Na Quan, Yongbiao Zhao, F. Pelayo Garcia de Arquer, Randy P. Sabatini, Grant Walters, Oleksandr Voznyy, Riccardo Comin, Yiying Li, James Z. Fan, Hairen Tan, Jun Pan, Mingjian Yuan, Osman M. Bakr, Zheng-Hong Lu, Dong Ha Kim, and Edward H. Sargent Nano Lett., Just Accepted Manuscript • Publication Date (Web): 05 May 2017 Downloaded from http://pubs.acs.org on May 5, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Tailoring the energy landscape in quasi-2D halide perovskites enables efficient green light emission Li Na Quan,1,2† Yongbiao Zhao,3† F. Pelayo García de Arquer,1† Randy Sabatini,1 Grant Walters,1 Oleksandr Voznyy,1 Riccardo Comin,1 Yiying Li,3 James Z. Fan,1 Hairen Tan,1 Jun Pan,4 Mingjian Yuan,1 Osman M. Bakr,4 Zhenghong Lu,3* Dong Ha Kim,2* Edward H. Sargent1* 1

Department of Electrical and Computer Engineering, University of Toronto, 10 King’s College Road, Toronto, Ontario, M5S 3G4, Canada.

2

Department of Chemistry and Nano Science, Ewha Woman’s University, 52, Ewhayeodae-gil, Seodaemun-gu, Seoul 03760, Korea.

3

Department of Materials Science and Engineering, University of Toronto, 184 College Street, Toronto, Ontario M5S 3E4, Canada. 4

Division of Materials Science and Engineering, King Abdullah University of Science and Technology, Thuwal 23955-6900, Kingdom of Saudi Arabia † These authors contributed equally to this work. E-mail: [email protected]; [email protected]; [email protected]

Organo-metal halide perovskites are a promising platform for optoelectronic applications in view of their excellent charge transport and bandgap tunability. However, their low photoluminescence quantum efficiencies, especially in low excitation regimes, limit their efficiency for light emission. Consequently, perovskite light emitting devices are operated under high injection, a regime under which the materials have so far been unstable. Here we show that, by concentrating photoexcited states into a small subpopulation of radiative domains, one can achieve a high quantum yield even at low excitation intensities. We tailor

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the composition of quasi-2D perovskites to direct the energy transfer into the lowestbandgap minority phase, and to do so faster than it is lost to non-radiative centres. The new material exhibits 60% photoluminescence quantum yield at excitation intensities as low as 1.8 mW/cm2, yielding a ratio of quantum yield to excitation intensity of 0.3 cm2/mW; this represents a two-orders of magnitude decrease in the excitation power required to reach high efficiency compared to the best prior reports. Using this strategy, we report LEDs with EQEs of 7.4% and a high luminescence of 8400 cd/m2. Keywords: Perovskites; Quasi-2D perovskites; Light-emitting diodes; Photoluminescence quantum yield; Energy transfer; Monte-Carlo Organo-metal halide perovskites (OHP) have emerged as a new class of optoelectronic materials for solar cells1 2 3, optically-pumped lasers 4 5 6, and light emitting diodes (LEDs) 7 8 9 10 11, 12

. However, their low exciton binding energy13 14 15, as well as their high electron and hole

mobilities, lead to poor photoluminescence quantum yields (PLQYs) at low excitation regimes 16

. In order to attain useful PLQYs, high excitation photon fluences must be supplied, and this

ultimately results in compromised photostability that limits light-emission applications17. Organic-inorganic hybrid perovskites have an ABX3 three-dimensional (3D) lattice framework that, through ion modification, can be spectrally tuned18. The low exciton binding energy in these systems results in a free-charge character in crystalline domains that slows radiative recombination15. Exciting the materials into a high injection regime is required to attain moderate light-emission efficiency. This, on top of the limited stability of these materials19, can produce unstable operation and fast degradation in light-emitting applications20. In contrast, lower dimensionality organic-inorganic hybrid perovskites are characterized by inorganic layers spaced using organic ligands21. This provides an excellent combination of


Page 2 of 25

Page 3 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

structural and photophysical properties, such as the ability to tune the material bandgap depending on the dimensionality22,

23

, increased exciton binding energy, and higher stability

compared to 3D counterparts19. Depending on the number of perovskite layers sandwiched between the organic linkers, these type of OHPs are termed 2D (e.g, a single PbBr2 layer) or quasi-2D (a distribution of layers with different thicknesses) perovskites. Low-dimensional perovskites are particularly attractive for visible light emission applications, and LEDs based on quasi-2D perovskites have been actively explored. However, these devices show low efficiency (below 1%) 24, 25. The radiative properties of quasi-2D perovskite systems based on PEA2(MA)n-1PbnI3n+1 (PEA=phenylethylammonium; MA=CH3NH3=methylammonium) have recently been reported for near-infrared LED applications26. The dimensionality of this material was tuned by varying the ratio of methylammonium iodide (MAI) to PEAI. This created a distribution of domains with varying numbers of perovskite layers centered on an average value . An impressive PLQY enhancement was achieved in the low regime. This was attributed to an inhomogeneous energy landscape that favoured funnelling of energy into domains with lower bandgap (higher dimensionality). A quantitative model of this phenomenon, together with a design strategy to unlock further advances in performance, remains to be reported. Herein we demonstrate the prerequisites for an efficient energy funnelling mechanism that gives rise to high photoluminescence yields in quasi-2D perovskite systems. We find that only when energy is funnelling from high bandgap domains into a small sub-population of lower bandgap domains does radiative recombination outcompete non-radiative recombination. Only then does the high charge concentration in these domains modulate charge-recombination


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 25

kinetics. In particular, we found that the distribution of these domains is crucial for achieving high PLQY and LED performance. Based on these findings, we then developed a fabrication strategy to control the domain distribution in quasi-2D perovskites. Specifically, we modulate the domain distribution in PEA2(MA)n-1PbnBr3n+1 perovskites through their composition, and solvent engineering during crystallization. In agreement with the model, only when the quasi-2D films are comprised of a distribution of different bandgap domains does highly efficient light-emission occur. The

energy-landscape-engineered

quasi-2D

perovskite

system

exhibits

high

photoluminescence yields at remarkably low excitation intensities (60% at 1.8 mW/cm2, corresponding to a ratio of quantum yield to excitation intensity of 0.3 cm2/mW). This represents an over-two-orders-of-magnitude decrease in the excitation density at which the quantum yield rises above the 50% mark when compared to previously-reported solution-processed perovskite PLQYs. Using these materials, we fabricated LED devices with external quantum efficiencies (EQEs) of 7.4% accompanied by high luminescence brightness (8400 cd/m2). These also feature notable improvements in stability compared to conventional MAPbBr3 perovskites. Results and discussion In conventional bulk 3D perovskites (Fig. 1a), excited charges diffuse faster than they recombine radiatively, ultimately becoming trapped at grain boundary defects. In lower dimensional systems, such as quasi-2D OHPs, a combination of free charges and excitons coexist in smaller grains, which increases the radiative recombination probability compared to the bulk-3D scenario (Fig. 1b). However, surface and interface defects can still dominate because of the smaller grain sizes and the extended interfaces27

28

. We hypothesized that, if the energy


Page 5 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

landscape in this material system was modulated in such a way that energy transfer could be fast enough to outpace trapping and subsequent non-radiative recombination, photoexcited states would be concentrated in small radiative domains where bimolecular radiative recombination would be enhanced because of the higher concentration (Fig. 1c). This would in turn result in high luminescence yields, even in low injection regimes. To gain insight into the conditions under which energy transfer is sufficient to enhance radiation efficiency, we developed a stochastic Monte Carlo model that accounted for charge recombination and transfer processes (Fig. 1d-e, Suppl. section). The dynamics of the excited population (N), following impulsive photoexcitation, can be described by the rate equation:

(

)

dN = − N A + BN + CN 2 + φ (t ) dt

(eq. 1)

where A, B, C are material-specific coefficients related respectively to Shockley-ReadHall (trap) recombination (1st order), bimolecular photoluminescence (2nd order), and Auger recombination (3rd order). ϕ(t) represents the rate of exciton generation per volume29. We first modelled the energy transfer and recombination yield of a flat energy landscape material (Fig. 1f, supplementary section). Under these conditions we observe no preferential energy accumulation (Fig. 1f inset) and, as a consequence, trapping becomes the dominant loss mechanism. Energy-landscape-engineered multidomain materials, on the other hand, display a significantly different behavior (Fig. 1g). For example, in a ternary system consisting of = {n1, n2, n3}, (see suppl. section for details), excitons sequentially funnelling from higher to lower bandgap domains, ultimately concentrating in final acceptor domains (n3) (Fig. 1g inset). In this


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

scenario, photon emission outcompetes trapping within the initial 10 ps (Fig. S10) as the strong charge concentration favours bimolecular radiative recombination (eq. 1). This downward energy cascade is expected to reduce the typically high optical pump powers and driving voltages required for efficient light emission in OHP systems due to the improved radiative efficiency. This is revealed by the computed photoluminescence quantum yield (PLQY) as a function of pump power intensity in different materials systems (Fig. 1h). Depending on the concentration coefficient f – where f denotes the effective increase of charge density in the final radiative domains compared to a flat-energy-landscape scenario – we observe a two orders-of-magnitude reduction in the pump power required to achieve high PLQY as f increases from 1 to 10. We conclude that energy funnelling in engineered landscapes can be exploited to improve the radiative properties of otherwise non-ideal materials, leading to efficient light emission achievable at lower driving powers. With these concepts in mind, we sought to define experimentally the energy landscape and energy flow dynamics in different OHP material systems. We synthesized quasi-2D bromide perovskites PEA2(MA)n-1PbnBr3n+1 at a judiciously selected stoichiometry (Supplementary Fig. S1)19,

26

. We employed different anti-solvents to control perovskite formation in a fast

crystallization single-step spin-coating method30. We took the view that the different boiling point of toluene (110°C) and chloroform (61°C) would result in different crystallization kinetics and resulting domain distributions. The high boiling point of toluene would favour the formation of more uniform domains (flat energy landscape) whereas chloroform, with faster evaporation, would lead to the formation of smaller clusters with different bandgaps (Supplementary Fig. S3 and S4).


Page 6 of 25

Page 7 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

We used ultrafast transient absorption spectroscopy (TA) measurements, to monitor the dynamics of energy transfer and thereby reveal the differences between these material configurations (Fig. 2). In the case of the flat energy landscape of standard quasi-2D perovskites, no energy migration is observed. A single bleach decay is recorded that coincides with the ground state bleach (GSB) and the excitonic peak obtained by linear absorption measurements (Fig. 2a-b). The extracted domain distribution is in this case dominated by a single domain component (Fig. 2c). Energy-landscape-engineered perovskites behave quite differently (Fig. 2d-f). Several GSBs are observed, and the decay kinetics of the different GSB peaks correlate well with the notion of energy transfer.31 The GSB maxima are sequentially delayed for lower energy features, indicative of domains that receive charges from (larger bandgap) domains. The energy transfer is observable in the first 10 ps, after which these points correspond, for each domain, to the time at which the population in each lower domain decays faster than it grows from energy funnelling. The decay kinetics are slower for lower energy features (acceptor domains). These facts, together, indicate that energy is funnelling from wider to narrower bandgap grains. The extracted domain distribution is in this case comprised by a set of components with different bandgaps that favour energy concentration into the emitting domain (Fig. 2e). Signatures of energy transfer in landscape-engineered perovskites are also evident in time-resolved PL emission studies, which show ultrafast (ps) and slow (ns) transfer components (Supplementary Fig. 9). We then sought to apply our Monte-Carlo model to fit the energy transfer kinetics in different landscape-engineered configurations to provide insights into the rates of the different processes involved (Fig. 3a). The different rise, maxima, and fall times for the various GSBs are evidence of energy transfer between the different domains which are in in good agreement with


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the theoretical predictions. Our model elucidates transfer rates on the order of ps-1 between the different perovskite domains, and estimates the instantaneous contributions of radiative, trap and Auger mechanisms (Supplementary Fig. S10). Photoexcited states’ lifetimes are shorter in widerbandgap grains, as these are the most likely to be transferred to narrower-bandgap grains over ~10 ps (Table 1). With these findings in mind, we then sought to further improve the final radiative efficiency, and therefore pursued control of the distribution of the different domains. We took the view that optimizing the concentration of the n = 5 emitting domains would allow a more efficient, graded funnelling, where the majority of the charges are concentrated in n = 5 domains rather than trapped or emitted in higher bandgap clusters before concentration. To do so we reduced the quantity of organic PEABr and MABr ligands involved in the perovskite synthesis, seeking to promote the formation of n = 5 domains. As revealed by transient absorption, the resulting films exhibited a different, broader distribution of n domains with a higher n = 5 contribution (Supplementary Fig. S8). We input the extracted domain distribution into our model, which yields good agreement with the experimental TA decay traces (Fig. 3b). The decay kinetics reveal a faster charge injection from the donor to the acceptor domains, which we attribute to the more graded domain distribution that increases the probability of sequential energy transfer. We then sought to assess experimentally how efficient energy transfer could lower the power at which high PLQY could be attained, and studied PLQY as a function of pump power for the different types of perovskite systems (Fig. 3c). As predicted, energy-landscapeengineered quasi-2D perovskites show a substantially higher PLQY compared to flat-landscape, bulk (MAPbBr3) and PEA2PbBr4 (2D) perovskites (Supplementary Fig. S7).


Page 8 of 25

Page 9 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Graded landscape =5 perovskites show a prominent PLQY of ~60 % at an excitation power density as low as 1.8 mW/cm2 (Fig. 3d). This yields a PLQY to pump power density ratio of 0.3 cm2/mW, a two-orders-of-magnitude improvement over prior perovskite systems (Table 2). We then translated the low operating powers of refined landscape-engineered quasi-2D perovskite materials into LED applications, and built LED devices employing =3 (engineered landscape), =5 (graded optimized landscape), and bulk MAPbBr3 (flat landscape) photoactive layers (Fig. 4a). Cross-sectional SEM indicates the presence of two uniform layers: a hole injection layer of PEDOT:PSS (Al4083) (~10 nm) coated with a dense perovskite emitter layer (~130 nm). 100 nm of TPBi was coated to provide an electron injection layer, followed by a LiF/Al (1/100 nm) electrode on top (Fig. 4b). Under forward bias, electrons and holes are injected from the cathode/electron injection layer (TPBi) and anode/hole injection layer (PEDOT:PSS) respectively, and are efficiently transferred to the perovskite, where they concentrate into the emitting domains to recombine radiatively. Ultraviolet photoemission spectroscopy (UPS) measurements were also performed to determine the valence band position and work function of the resulting perovskites (Supplementary Fig. 6). Flat landscape bulk perovskite devices exhibit low performance with a 0.8% EQE and 33.7 cd/m2 luminance (Fig. 4c and d). This is expected in view of their lower PLQY (Fig. 3f). Devices based on landscape-engineered quasi-2D perovskites, on the other hand, show improved optoelectronic characteristics. =3 devices exhibit reduced turn-on voltages and an order-ofmagnitude increase in EQE and luminance (up to 4.8% and 1000 cd/m2) (Fig. 4c and d). Graded =5 devices display a significant enhancement in EQE (7.4%) and luminance (8400 cd/m2) (Fig. S11 and S13). The EL spectra of these devices closely match the PL spectra, preserving a


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 25

narrowband emission (Supplementary Fig. S12). Large area devices (6.54 cm2) based on graded =5 perovskites were also fabricated (Fig. 4d inset) and show high brightness with strong optical uniformity. The low device operational stability may due to the material photostability and interaction of perovskite with the device interfacial layers (Fig. S14). Conclusion We demonstrated the conditions required for efficient radiative recombination in low dimensional perovskite systems. We found that when the energy landscape in the perovskite is modulated in such a way that energy transfer outpaces non-radiative recombination, concentrating charges into a subpopulation of radiative domains, efficient light emission take place. We devised a perovskite synthesis and crystallization method whereby domain distribution could be controlled. By optimizing the domain distribution, we achieved a record PLQY of 60% at low excitation fluences (1.8 mW/cm2) for solid films at green wavelengths. This represents the highest maximum yield-to-driving-power reported for organo-metal halide perovskite materials (0.3 cm2/mW). We translated these findings into LED devices with high EQEs (7.4%) and high luminescence (8400 cd/m2) at low threshold voltages. Materials and Methods Perovskite film fabrication. The quasi-2D and MAPbBr3 perovskites precursors were prepared by dissolving specific stoichiometric quantities of lead bromide (PbBr2, 99.98% Alfa-Aesar), methylammonium bromide (MABr, Dyesol) and phenylethylammonium bromide (PEABr, Dyesol) in DMSO solvent. The resulting solution was filtered using a PTFE syringe filter (0.2 µm) before deposition. The precursor solution was coated onto the substrate via a consecutive two-step spin coating process at 1000 r.p.m. and 5000 r.p.m. for 10 and 60 seconds, respectively. During the


Page 11 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

second spin step, 100 µL of chloroform (or toluene) were deposited onto the substrate. The resulting films were then annealed at 90ºC for 5 minutes for better crystallization. Moreover, the time of dripping anti-solvent (before the evaporation of DMSO) is essential to get highly quality and better crystallized film. In order to assess the crystal structure and layered nature of the resulting films, we measured film X-ray diffraction (XRD) of pure phase MAPbBr3 and quasi-2D perovskites with =3 and =5 (Supplementary Fig. 2). The XRD patterns of MAPbBr3 films reveal Bragg reflections at 15.23° and 30.44° that can be indexed to the (100) and (200) planes. This indicates a high purity cubic phase of MAPbBr3 perovskite. Signatures of layered perovskite structure were obtained at low-diffraction angles (2θ < 10°) in =3 perovskite films. Similar reflection patterns for =5 and MAPbBr3 indicate a preferential orientation of the grains on the substrate. Interestingly, quasi-2D perovskites with = 3 and = 5 show extremely small grain sizes, on the order of 10 nm, and high smoothness, below 10 nm rms. MAPbBr3 perovskites, on the other hand, consistently present larger grains (>50 nm) and high surface roughness (above 50 nm rms.) (Supplementary Fig.3). Grain sizes calculated from XRD peaks FWHM are consistent with AFM results. Light-Emitting Diodes fabrication. A PEDOT:PSS (CleviosTM PVP Al4083) layer was spin-coated on oxygen-plasma-treated patterned ITO-coated glass substrates, then annealed on a hot plate at 150ºC for 20 minutes in air. Perovskite precursor solutions were spin-coated onto the PEDOT:PSS via the two-step fast crystallization spin-coating method that was described above. TPBi (100 nm) and LiF/Al electrodes (1 nm/100 nm) were deposited using a thermal evaporation system through a shadow mask under a high vacuum of less than 10-4 Pa. The device active area was 12.2 mm2 as defined by the overlapping area of the ITO and Al electrodes. Unpatterned ITO substrates (1 by 1 inch)


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 25

were used for the large-area devices. All devices were tested under ambient condition with encapsulated devices.

AFM measurements. Atomic force microscopy (AFM) was used to characterize sample morphology. An Asylum Research Cypher AFM was operated in AC mode in air to obtain topographical and phase images. ASYELEC-02 silicon probes with titanium-iridium coatings from Asylum Research were used for all imaging. The typical spring constant of the probes is 42 N/m. XRD measurements. XRD measurement on oriented films were conducted on a Panalytical X’Pert Pro diffractometer with a Bragg-Brentano geometry and PLXCEL 1D detector equipped with a nickel filter. UPS measurements. UPS spectra of the perovksite films were measured on Au coated substrate. Photoelectron spectroscopy was performed in a PHI5500 Multi-Technique system using non-monochromatized He-Iߙ radiation (UPS) (hv=21.22ev). All work function and valence-band measurement were performed at a takeoff angle of 88º, with chamber pressure near 10-9 Torr. Photoluminescence (PL) measurements. Photoluminescence (PL) measurements were performed using a Horiba Fluorolog system. Steady-state PL was collected by illuminating the samples with a monochromatized Xe lamp. Transient PL was acquired with a Time Correlated Single Photon Counting detector and a pulsed UV laser diode ( ߣ = 375 nm). The instrument response function provides an overall time resolution of Δ‫~ݐ‬0.13 ns. Time-resolved emission spectra were mapped by collecting individual transient PL traces at different emission wavelengths. Absolute PL quantum yield (PLQY)


Page 13 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

measurements were done by coupling a Quanta-Phi integrating sphere to the Fluorolog system with optical fiber bundles. The measurements were done in accordance with published methods27. Both excitation and emission spectra were collected for the three cases of the sample directly illuminated by the excitation beam path in the integrating sphere, the sample offset within the integrating sphere from the beam path, and the empty sphere itself. The PLQY measurements were done by setting the Fluorolog to an excitation wavelength of 400 nm and to have a 5 nm bandpass for both the excitation and emission slits. With these settings, the ensuing spectra had high signal to noise ratios and delivered an excitation intensity in a range of 1-30 mW/cm2 to the sample. Excitation intensity spectra were collected with a calibrated neutral density filter with known transmission placed after the integrating sphere. A Newport white light source was used to calibrate the detector and integrating sphere for spectral variance. PL spectra were collected as a function of excitation intensity by varying the slit width on the Fluorolog monochromator. The excitation intensity was calculated by measuring the power with an Ophir LaserStar Dual Channel Power and energy meter and calculating the beam area through the known dispersion relations for the monochromator. Transient absorption (TA) measurements. A regeneratively amplified Yb:KGW laser at a 5 kHz repetition rate (Light Conversion, Pharos) was used to generate femtosecond laser pulses, and a pulse picker was used to lower the frequency to 1 kHz. A portion of the 1030 nm fundamental was sent into an optical bench (Ultrafast, Helios), where it passed through a retroreflector, and was then focused into a calcium fluoride crystal, translated at 1 mm/s, to create the white light continuum probe. An optical parametric amplifier (Light Conversion, Orpheus) was used to generate the 350 nm pump pulse by upconversion of the fundamental wavelength. This was then sent into the optical bench and


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 25

was chopped at 500 Hz. Both the pump and probe were sent to the sample, with the time delay adjusted by changing the pathlength of the probe (time resolution ~ 350 fs). The probe pulse was then collected by a CCD after dispersion by a grating spectrograph (Ultrafast). Pump fluences were kept at 20 µJ/cm2. Kinetic traces were fit to the convolution of the instrument response and a sum of exponential decays. Time zero was allowed to vary with wavelength to account for the chirp of the probe. All TA measurement results were plotted based on the ref

28

. The relative

presence of each domain (Figures 2 and 3) was quantified by the amplitude of the transient absorption signal at t = 0 as:

ρni =

∫ ∆AdE

ni

∫ ∆AdE

(eq. 2)

n

Device characterization The luminance (L) - current density (J) – voltage (V) characteristics were collected by using a HP4140B picoammeter, a calibrated luminance meter (Konic Minolta LS-110). The absolute EL power spectra of the devices were collected with an integrating sphere and an Ocean Optics USB4000 spectrometer by mounting the devices on the wall of the integrating sphere. The EQE was then calculated through the measured absolute power spectra and current density. Calculations The procedure describing the applied stochastic Monte Carlo simulations is detailed in suppl. section S1. The ratio of excitons and free carriers in perovskite systems has been calculated based on ref 29. The PLQY as a function of power and funnelling factor was obtained from:

PLQY ( f , Ppump ) =

k rad

k rad + ktrap + k Auger


(eq. 3)

Page 15 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

where for simplicity the kAuger term has been included. Flat and engineered energy landscape modeling: The energy evolution in plain landscape materials was evaluated according to the next parameters: A = 2·109 s-1, B=108 cm3 s-1 and C = 1.6·106 cm6 s-1, excited using a 150 fs pump excitation of 20 µJ/cm2 . For engineered energy landscape films, the individual recombination rates of the ni domains have been kept equal to those typical of the single material32. A transfer rate of ktransfer= 1011 s-1 was employed and the contribution of each domain weighted by w = (0.6, 0.3, 0.1) and w = (0.2, 0.5, 0.3) for the graded sample.


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 25

Figure 1. Tailoring the energy landscape in quasi-2D halide perovskites for efficient light emission. (a) In bulk-3D perovskites, the high diffusivity of charges militate against radiative recombination. (b) In quasi-2D perovskites, free-charges and excitons coexist. A competition between radiative and non-radiative recombination, exists determined by density of defects and on the density of photoexcited charges and excitons. (c) In energy-landscape-engineerd quasi-2D perovskites, energy funnelling occurs faster than non-radiative recombination. As a consequence, photoexcited states are strongly concentrated in a subpopulation of domains, thereby increasing the probability of bimolecular radiative recombination. (d) In a flat energy landscape material, charges can either be trapped (with a rate ktrap), or lost via Auger (kAuger) or radiative (krad)


Page 17 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

recombination. The presence of trap states within, and at the boundaries, of the different grains, results in significant non-radiative recombination. (e) This limiting process can be overcome provided energy is funnelling to lower-bandgap domains before non-radiative recombination occurs. (f) Monte-Carlo simulation of the time evolution of photoexcited states after pumping in a flat energy landscape system. Inset: distribution of excited states within the organohalide perovskite matrix after the initial 10 ps. After initial excitation only trapping events are recorded. (g) In energy-landscape-engineered systems, energy transfer outcompetes non-radiative processes. The photoexcited states are fully transferred from n1 to n2 and n3 domains within the first 100 ps. Inset: distribution of excited states after the initial 10 ps. An increasing contribution of radiative events is recorded upon exciton funnelling (Fig S1). (h) The strong energy concentration in energy-landscape-engineered quasi-2D systems enabler high photoluminescence quantum yield at much lower pump intensities.


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 25

Figure 2 | Controlling the energy landscape for quasi-2D perovskites through the crystallization process. Transient absorption spectra of flat (a-c) and (d-f) energy-landscapeengineered =3 quasi-2D perovskite films. The relative presence of different n domains was extracted from the amplitude of the ground state bleach peaks (see methods). In the case of a flat energy landscape, n=5 dominates. Energy-landscape-engineered films, instead, consist of a set of different domains where energy funnels into the final n=5 domain.


Page 19 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 3 | Optimizing energy transfer in energy-landscape-engineered quasi-2D perovskite systems. (a-d) Contribution of different domains in energy-landscape-engineered and gradedenergy-landscape quasi-2D perovskite films as estimated from transient absorption spectroscopy (c-d) Experimental and predicted time dependent transient absorption traces for engineered =3 (c) and graded =5 (d) films for different wavelengths. (e) PLQY as a function of pump power density for different material systems, showcasing the benefits of energy-landscapeengineered quasi-2D perovskites. (f) PLQY for different quasi-2D perovskite configurations.


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 25

Figure 4 | Energy-landscape-engineered quasi-2D perovskites for highly efficient LEDs (a) Band structure of perovskites with different perovskites and energy band alignment of LED device. (b) Cross-sectional SEM image of LED device (artificially colored). (c) EQE versus applied voltage characteristics of the device with different perovskites. (d) Summary of the J-VL characteristics in different perovskites based device.


Page 21 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Table 1. TA (Experimental) lifetime =3

=5

430 nm

460 nm

480 nm

490 nm

460 nm

500 nm

5.7 ps

16 ps

30.7 ps

70.5 ps

3.5 ps

6.07 ps

42 ps

198 ps

169 ps

1 ns

26 ps

3 ns

st

1 exponential 2nd exponential

Table 2. PLQY, pump power (Ppump) and PLQY-to-Ppump Reference 15 7 7 10 Flat energy landscape (This work) Engineered landscape This work

Year 2014 2014 2014 2015

Regime NIR NIR Visible Visible

PLQY (%) 50 26 7 36

Ppump (mW cm-2) 100 333 1000

PLQY/ Ppump 0.005 0.00078 0.00036

Perovskite MAPbIxCl3-x MAPbIxCl3-x MAPbBr3 MAPbBr3

2016

Visible

12

1.8

0.06

PEA2(MA)4Pb5Br16

2016

Visible

60

1.8

0.33

PEA2(MA)4Pb5Br16


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Graphic


Page 22 of 25

Page 23 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Supporting Information Description of Monte Carlo modelling of recombination kinetics, unit cell structure of perovskites, XRD, AFM, Absorption and PL, UPS spectra, Transient absorption data, Transient PL, LED characteristics, EL spectra and LED stability. This material is available free of charge via the Internet at http://pubs.acs.org. Author Contributions L.Q., Y.Z., F.P.G.A and E.H.S. conceived the idea and proposed the experimental design. L.Q., F.P.G.A., R.S., G.W., O.V., R.C., Y.L., J.F., H.T., P.J., M.Y., and D.H.K. performed and analyzed XRD, UV absorption, PL lifetime, transient absorption and UPS, SEM measurements. L.Q. Y.Z. performed the device fabrication. F.P.G.A performed the simulation. L.Q., Y. Z. and Z. Lu tested the devices. L.Q., F.P.G.A., O.V. and E.H.S. co-wrote the manuscript. All authors read and commented on the manuscript. Notes The authors declare no competing financial interests. Acknowledgements This publication is based in part on work supported by Award KUS-11-009-21, made by King Abdullah University of Science and Technology (KAUST), by the Ontario Research Fund Research Excellence Program, and by the Natural Sciences and Engineering Research Council (NSERC) of Canada. L. N. Quan and D. H. Kim acknowledge the financial support by National Research

Foundation

of

Korea

Grant

funded

by

the

Korean

Government

(2014R1A2A1A09005656; 2015M1A2A2058365). F. P. García de Arquer acknowledges financial support from the Connaught fund.


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 25

References (1) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Science 2012, 338, 643-647. (2) Burschka, J.; Pellet, N.; Moon, S.-J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Grätzel, M. Nature 2013, 499, 316-319. (3) Jeon, N. J.; Noh, J. H.; Yang, W. S.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I. Nature 2015, 517, 476-480. (4) Xing, G.; Mathews, N.; Lim, S. S.; Yantara, N.; Liu, X.; Sabba, D.; Grätzel, M.; Mhaisalkar, S.; Sum, T. C. Nat. Mater. 2014, 13, 476-480. (5) Sutherland, B. R.; Hoogland, S.; Adachi, M. M.; Wong, C. T.; Sargent, E. H. ACS nano 2014, 8, 10947-10952. (6) Zhu, H.; Fu, Y.; Meng, F.; Wu, X.; Gong, Z.; Ding, Q.; Gustafsson, M. V.; Trinh, M. T.; Jin, S.; Zhu, X. Nat. Mat. 2015, 14, 636-642. (7) Tan, Z.-K.; Moghaddam, R. S.; Lai, M. L.; Docampo, P.; Higler, R.; Deschler, F.; Price, M.; Sadhanala, A.; Pazos, L. M.; Credgington, D. Nat. Nanotechnol. 2014, 9, 687-692. (8) Hoye, R. L.; Chua, M. R.; Musselman, K. P.; Li, G.; Lai, M. L.; Tan, Z. K.; Greenham, N. C.; MacManus‐Driscoll, J. L.; Friend, R. H.; Credgington, D. Adv. Mat. 2015, 27, 1414-1419. (9) Kim, Y. H.; Cho, H.; Heo, J. H.; Kim, T. S.; Myoung, N.; Lee, C. L.; Im, S. H.; Lee, T. W. Adv. Mat. 2015, 27, 1248-1254. (10) Cho, H.; Jeong, S.-H.; Park, M.-H.; Kim, Y.-H.; Wolf, C.; Lee, C.-L.; Heo, J. H.; Sadhanala, A.; Myoung, N.; Yoo, S. Science 2015, 350, 1222-1225. (11) Li, G.; Tan, Z.-K.; Di, D.; Lai, M. L.; Jiang, L.; Lim, J. H.-W.; Friend, R. H.; Greenham, N. C. Nano Lett. 2015, 15, 2640-2644. (12) Xiao, Z.; Kerner, R. A.; Zhao, L.; Tran, N. L.; Lee, K. M.; Koh, T.-W.; Scholes, G. D.; Rand, B. P. Nat. Photonics 2017, 11, 108-115. (13) Hong, X.; Ishihara, T.; Nurmikko, A. V. Phys. Rev. B 1992, 45, 6961. (14) Tanaka, K.; Takahashi, T.; Ban, T.; Kondo, T.; Uchida, K.; Miura, N. Solid. State. Commun. 2003, 127, 619-623. (15) Yang, Y.; Yang, M.; Li, Z.; Crisp, R.; Zhu, K.; Beard, M. C. J. Phys. Chem. Lett. 2015, 6, 4688-4692. (16) Deschler, F.; Price, M.; Pathak, S.; Klintberg, L. E.; Jarausch, D.-D.; Higler, R.; Hüttner, S.; Leijtens, T.; Stranks, S. D.; Snaith, H. J. J. Phys. Chem. Lett. 2014, 5, 1421-1426. (17) Stranks, S. D.; Burlakov, V. M.; Leijtens, T.; Ball, J. M.; Goriely, A.; Snaith, H. J. Phys. Rev. Applied 2014, 2, 034007. (18) Saparov, B.; Mitzi, D. B. Chem. Rev. 2016, 116, 4558-96. (19) Quan, L. N.; Yuan, M.; Comin, R.; Voznyy, O.; Beauregard, E. M.; Hoogland, S.; Buin, A.; Kirmani, A. R.; Zhao, K.; Amassian, A. J. Am. Chem. Soc. 2016, 138, 2649-2655. (20) Yu, J. C.; Kim, D. W.; Kim, D. B.; Jung, E. D.; Park, J. H.; Lee, A. Y.; Lee, B. R.; Di Nuzzo, D.; Friend, R. H.; Song, M. H. Adv. Mat. 2016, 28, 6906-6913. (21) Mitzi, D. B.; Feild, C.; Harrison, W.; Guloy, A. Nature 1994, 369, 467-469. (22) Sichert, J. A.; Tong, Y.; Mutz, N.; Vollmer, M.; Fischer, S.; Milowska, K. Z.; García Cortadella, R.; Nickel, B.; Cardenas-Daw, C.; Stolarczyk, J. K. Nano lett. 2015, 15, 6521-6527. (23) Milot, R. L.; Sutton, R. J.; Eperon, G. E.; Haghighirad, A.-A.; Martinez Hardigree, J. F.; Miranda, L.; Snaith, H. J.; Johnston, M. B.; Herz, L. M. Nano Lett. 2016, 16, 7001-7007.


Page 25 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(24) Wang, N.; Cheng, L.; Ge, R.; Zhang, S.; Miao, Y.; Zou, W.; Yi, C.; Sun, Y.; Cao, Y.; Yang, R. Nat. Photonics 2016, 10, 699-704. (25) Byun, J.; Cho, H.; Wolf, C.; Jang, M.; Sadhanala, A.; Friend, R. H.; Yang, H.; Lee, T. W. Adv. Mat. 2016, 28, 7515-7520. (26) Yuan, M.; Quan, L. N.; Comin, R.; Walters, G.; Sabatini, R.; Voznyy, O.; Hoogland, S.; Zhao, Y.; Beauregard, E. M.; Kanjanaboos, P. Nat. Nanotechnol. 2016, 11, 872-877. (27) Wu, X.; Trinh, M. T.; Niesner, D.; Zhu, H.; Norman, Z.; Owen, J. S.; Yaffe, O.; Kudisch, B. J.; Zhu, X.-Y. J. Am. Chem. Soc. 2015, 137, 2089-2096. (28) Sum, T. C.; Mathews, N.; Xing, G.; Lim, S. S.; Chong, W. K.; Giovanni, D.; Dewi, H. A. Acc. Chem. Res. 2016, 49, 294-302. (29) Manser, J. S.; Kamat, P. V. Nat. Photonics 2014, 8, 737-743. (30) Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Seok, S. I. Nat. Mat. 2014, 13, 897-903. (31) Berera, R.; van Grondelle, R.; Kennis, J. T. Photosyn. Res. 2009, 101, 105-118. (32) Rowland, C. E.; Fedin, I.; Zhang, H.; Gray, S. K.; Govorov, A. O.; Talapin, D. V.; Schaller, R. D. Nat. Mat. 2015, 14, 484-489.


Tailoring the Energy Landscape in Quasi-2D Halide Perovskites

Recommend Documents